The Designer's Guide Community Forum

The Designer's Guide Community Forum
https://designers-guide.org/forum/YaBB.pl
Other CAD Tools >> Physical Verification, Extraction and Analysis >> Calibre view wrong MOS properties after PEX extraction
https://designers-guide.org/forum/YaBB.pl?num=1287768249

Message started by .matteo on Oct 22^nd, 2010, 10:24am

Title: Calibre view wrong MOS properties after PEX extraction
Post by .matteo on Oct 22^nd, 2010, 10:24am

Hi everybody,
I'm using Calibre PEX to extract one nMOS device (W=5,L=Lmin,NF=2,mult=1) after passing the DRC and LVS. I select the "calibreview" tick in the PEX Outputs options + noRC, noL.
PEX runs without errors and outputs the attached pxi file; please have a look at the whole file attached if you want.

Code:

...
mgc_rve_cell_start "tst_nMos" "G" "D" "gnd"
mr_pi "nlvtlp_rf" "M0" '( "D" "G" "gnd" "gnd") '( ("l" 0.06) ("w" 5) ("nfing" 2) ("ngcon" 1) ("mult" 0.5) ("srcefirst" 1) ("as" 0.88) ("ad" 0.495) ("ps" 5.704) ("pd" 0.396) ("po2act" 0.250439) ("$x" -1250) ("$y" -159) ("$d" 241) ("lpe" 0) ) '(-1.25 -0.159)
mr_pi "nlvtlp_rf" "M0@2" '( "D" "G" "gnd" "gnd") '( ("l" 0.06) ("w" 5) ("nfing" 2) ("ngcon" 1) ("mult" 0.5) ("srcefirst" 1) ("as" 0.88) ("ad" 0.495) ("ps" 5.704) ("pd" 0.396) ("po2act" 0.250439) ("$x" -1250) ("$y" 101) ("$d" 241) ("lpe" 0) ) '(-1.25 0.101)
mgc_rve_cell_end
...

As you can see, PEX extract the MOS correctly aka two MOS devices each with W=5,L=Lmin,NF=2,mult=0.5.

The problem comes when I create the calibre view from the calibre view setup window that pops up. The view that is created contains two MOS devices with W=5,L=Lmin,NF=2,mult=1 ..that is wrong!!

How can I tell calibre to create the view correctly?
Maybe modifying the cellmap file or setting up something in the "reset properties" field in the calibre view setup?

Really hope somebody can help me out with this

Thanks
.matteo

Title: Re: Calibre view wrong MOS properties after PEX extraction
Post by Alexandar on Oct 28^th, 2010, 4:59am

If you don't use the multiplier in your schematics, and use an arrayed version of your instance instead, you can bypass this problem. I've seen this problem happening with transistors and resistors. Somehow the PEX does not take into account the 'm'.

Title: Re: Calibre view wrong MOS properties after PEX extraction
Post by Paul H on Nov 8^th, 2010, 6:30am

Hi matteo

I don't know what technology your using but are you sure it's right to have a mult=0.5 - this is half a device !

Calibre will extract each gate over active as an individual device (with associated parasitics) so the mult parameter will need to be set at 1, disregarding any setting in the corresponding schematic.

To do this you need to add a statement to the nlvtlp_rf device section of the calibre view mapping file which will set the mult parameter of the device created in the calibreview (this is the dfII schematic of the extracted layout) to a required value, in this case 1.

mgc_device( "p"
'( "<cdn lib name>" <cell name> symbol)
'( (b B) (d D) (g G) (s S) )
'( (l l) (w w) (mult m 1) )
)

The first parameter name in the pairings is the Calibre name and the second is the matching CDF property name.

This is a question that you should have had answered by your Mentor support team. To get the fastest response for your Calibre queries use Supportnet (supportnet.mentor.com) which will get you direct access to Mentor customer support.

Thanks
Paul